 Welcome to the search testing phase of how to conduct a systematic search. This is useful for doing any kind of search that would be involved in a systematic review, a meta-analysis, a scoping review, a rapid review, an umbrella review, or any other type of literature review that you might want to use a systematic search for like an integrative review or a narrative review. So I've assembled quite a body of keywords here, and this is for a systematic review that would be looking at the Pico question. What is the impact of early childhood exposure to highway pollution and the link to developing ADHD or presentation of ADHD later in life? And this is based on some systematic reviews that I've seen before. There's a lot available for this. So what I've done is I've assembled a pretty big broad body of keywords here. And so what I'm going to do is I'm going to move into a testing phase. So what that involves is taking some of our terms here, and then I'm going to use them in PubMed to start doing some testing here. So we have these big long strings conceptual keywords here. So we have one for our different types of pollution or impact factors that might impact pollution. We have the ways the exposure happens. So looking at exposure, early life, et cetera. And then we have the disorder that we want to look at, which is ADHD and some different terminology for that. So I want to move into testing this ADHD string, which I gathered from a bunch of things using my match terms and stuff. So what I'm going to do is I'm actually going to do some testing in PubMed because PubMed has some tools that can help us test our search. And this works really well if you're using a biomedical search. But if you're not doing a biomedical search, choose a database that works well for the subject that you're using. So if you're doing something in education, use Eric, for example. So we'll need to go back to PubMed. And then what I'm going to do here is instead of just copying and pasting there, I'm going to go to an advanced search because what we want to do is specify what fields we want to search. We're not going to do an all field search because that does some things that we don't necessarily want it to do. It does what's called automatic term mapping, which maps our terms to things we don't necessarily have control over. But if we search the titles and abstracts for our concepts, for our terms, that might actually lead us to something more productive, more useful because they tend to be more descriptive in titles and abstracts when we're searching for biomedical literature in particular. So I'm going to add each of these and it adds that field tag right here, that title, abstract field tag. That's going to add that to that and then start to shape our search for us. So I'm going to run that and we can see that we get a lot of results for this. Now, if I go back to my advanced search, we can see that it got all these search results and we can see that it found all of my terms. If it didn't find my terms under the details here, it would show me which terms it didn't find. So that helps me know that these are all good terms that I want to use. And also this title, abstract, is something that I can copy and paste and use in my search. The other thing that I want to do to start testing the search is I also want to add a mesh term to the title and abstract. So we need to go back and grab that ADHD mesh term as well. So what I'm going to do is I'm going to start back up PubMed, go into the mesh database, type in ADHD and add this to my search builder and search PubMed. And that gives us 32,000 results. Now, if I go back into my advanced search, then I can combine the mesh term with or the title, abstract search. And what this does is it's going to try to find any instance of these situations coming up. You can see I've also included some signature ADHD symptoms to see if those might be showing up as presentations too. And then with that search, we get a much wider pool of ADHD types of resources. So this is how we're starting to do our search test. I can then go back into advanced search and start to test some of my other groups of searching. So right here, I have this list of exposure criteria. So these are the ways that we want to look at the exposure or the influence. And so we can copy and paste that in there and then look for title, abstracts and then add that in and then run that search too. So again, we get a lot there. Go to advanced, make sure it's finding all of those terms and it sure is. Go back to our plain text sheet and then do the same for our air pollution our air pollution concept or roadway proximity. There's a lot of ways to refer to that. And again, I'm gonna do a title abstract search because that's where the most relevant information is going to be. And then I'm gonna run those searches. Oh, so it actually didn't find something here. So this is why we run the search test is you can go in and see what it didn't find. And so it flagged us here. And it looks like, oh, you know what? I put a star before benzene and then I put an O at the beginning or at the end. So that star before benzene, I was trying to look for things that end in benzene and that is not something that it's going to do here. So what I'm gonna end up doing is I'm just gonna look for benzene or benzines. So I'll probably end up changing the search a little bit. So what I can do is just grab this part, copy it, paste it in here. And then I can go through trying to find benzene and I'm gonna look for benzines instead and then make sure I don't have that act, that or at the end. And then that's a much better result. And then that's our basic search there. And then what we can start to do is start to combine our searches. So we've got this one right here. We can get rid of it because it's no good. And then we can start to do some other things. So we can grab some, so now that we have these title abstract phrases ready to go in PubMed, we can start to copy those and add those to our spreadsheet here. So we have that one and then we have that one. And then we'll grab this last one here. And so we have basically our PubMed search coming together right now in the proper search syntax. So what this is doing is it's telling the search engine exactly where to search for these terms. Now, instead of going through and trying to find the best terms for all of these different searches instead, I've actually gone ahead and made them before from another search. So for this one I have, so we have all of our different airhouse pollution. So we just wanna group all those things together, our mesh terms and our title abstract searches of the same concepts need to be grouped together. So we also know that or we know what's there too. Similarly, I have another batch from a similar search right here. I can copy that, I will paste that into this one here or that. So this way we have all of our searches ready to go and then we can start to make the search work and run a test. So I have my ADHD search here and I can add that and that's ready to go. And what that does is it put parentheses on it because that groups everything together and says like an order of operations in math, this is our search and we're ready to go. Then the next thing I can do is up here, I can start to copy and paste my other search strings. So we can take this roadway proxima, copy this, just that there and then add that with and because we're gonna try to find the overlap between each of these three things. And find where these three concept groups overlap together. And then we'll do that here with the title abstracts of epigenetic factors and then run the search. And this is where we can see that we get 218 results which is pretty good. Now I have these over in my Zotero library, I'm gonna drag and drop that here. I have a list of six articles that I have decided to use as what I'm calling my development set and what these are are these are articles that I wanna start looking for in my results. So these are different articles that I wanna start looking for within my search results. If we have specific titles that we can bring in, we can take this particular search and then we can add in, say for authors, we wanna try to find Pereira, for example. This is someone who we know does research on this area. So we can draw in her name and run that search and see, okay, we've got Pereira writing on this. So we know we're doing a pretty good type of search here. We can see early life exposure. I think that's exactly this one right here. So we managed to find exactly one of our articles that we had in the development set, which is even better. So this is how you can run your test search for doing a systematic review. And so that's just how you format it for one database. For each database, you need to do something different. So we go into, for example, Web of Science or something. Science doesn't do the same things that our other databases do. We have to kinda translate this search for there. So that's why we keep each of these phrases right here. So we can just grab ADHD. Instead of searching all fields, if you do topic that searches the title, the abstract, and then any author added keywords. So again, searching topic here can be really useful to our systematic review. And then if we add in our next thing. So you might even find keywords as you're doing this process that you end up adding to your search. So this is why this process is really important is because you might encounter things as you're searching that could be added to your search. So we get 812 here because this is a much larger database. So I'm getting a lot of not super relevant stuff here, which makes me wonder, I wonder if I need to change the Benzines again because sometimes that Benzines gets really weird. See, we get rats. So we wanna kinda try to keep on focusing this a little bit. Oh, I bet it's add is throwing us off because that could be, they're giving us results for like adding and stuff. Let's try, oh, see, there we go, changed everything. So that's why you wanna go through and start testing your searches because ADAT might not mean something in every database. Might not mean the same thing in each database you use. You can even see right up here what they're suggesting with this ads. What that's doing is they're trying to add in terminology to our search that is not useful or relevant to us. So that's part of what's creating this problem. And so that's kinda how you can go about doing some problem solving during your test searching phase.